√RAN “wander, stray, meander, go on an uncertain course, go aside from a course (commanded or self-chosen); err”
This root first appeared as ᴹ√RAN “wander, stray” in The Etymologies of the 1930s with derivatives like ᴹQ. ranya-/N. rhenia- “to stray”, ᴹQ. ránen/N. rhaun “errant”, and ᴹQ. Rana/N. Rhân as names for the Moon (Ety/RAN). These Moon-names also appeared in earlier writings but without clear etymologies (LT1/192; GL/64). Such Moon names continued to appear in later writings, for example: Q. Rána “Wayward” (S/99). The root √RAN itself was mentioned quite frequently in Tolkien’s later writings with glosses like “wander, stray” (PE17/182), “wander, stray, go on uncertain course” (VT42/12) and “err, go aside from a course (commanded or self-chosen)” (PE17/78).
ᴱ√RAPA “drag, pull”
A root in the Qenya Lexicon of the 1910s with a series of glosses “handle hastily, snatch, grab” >> “tear” >> “drag, pull” but no derivatives (QL/79). The similar form G. drab “labour, toil, irksome work” and other derived words from the contemporaneous Gnomish Lexicon might be related (GL/30), or could represent a distinct root *ᴱ√DRAP “labour, toil”. ᴱN. drauth “labour” and draust “work” from the Early Noldorin Word-lists from the 1920s seem to be the last published iteration of these forms (PE13/142, 156). The original QL gloss “handle hastily, snatch, grab” might have reemerged in ᴹ√RAPH “snatch” from the 1940s; see the entry √RAPH for further details.
Neo-Eldarin: For purposes of Neo-Eldarin, I think it is worth positing a Neo-Root ᴺ√(D)RAP “toil” to salvage Gnomish drab- words.
√RAPH “seize, grab, [ᴹ√] snatch”
The root ᴹ√RAPH “snatch” itself first appeared in the Quenya Verbal System with a derived verb of ᴹQ. raf- of the same meaning (PE22/102), but it might have been a reemergence of the very first (deleted) gloss of the early root ᴱ√RAPA the Qenya Lexicon of the 1910s: “handle hastily, snatch, grab” (QL/79). As for √RAPH, it reappeared in green-ink revisions to the Outline of Phonology (OP2) from around 1970 with the gloss “seize, grab” and a derivative Q. arpo “seizer, thief” (PE19/89 and note #101). On the same page but from the originally layer of composition from the early 1950s there is the verb Q. raf- “wave, brandish” which may be related, though if so it must represent some semantic drift (PE19/89).
√RAS “horn; [ᴹ√] stick up”
This root first appeared as ᴹ√RAS “stick up (intr.)” in The Etymologies of the 1930s with derivatives like ᴹQ. rasse and N. rhas or rhasg “horn (especially on living animal, but also applied to mountains)” (Ety/RAS; EtyAC/RAS). It reappeared as ᴹ√RASA “stick up” on an rejected page of roots in the Quenya Verbal System from the 1940s (PE22/127). Finally, √RAS “horn” appeared in Common Eldarin: Noun Structure of the early 1950s, but that was merely the last appearance of the root in Tolkien’s published writings. Q. rassë and S. rass “horn” continued to appear regularly as an element in mountain names in the 1950s and 60s.
ᴹ√RASAT “twelve”
The root ᴹ√RÁSAT “twelve” appeared in The Etymologies of the 1930s, and students of Elvish long supposed that its unattested derivative *rasta was the Quenya word for “twelve”. In later publications, however, the Quenya word for “twelve” was given as yunquë (VT47/41), derived from primitive ✶yūnekē.
This word yunquë dates back to Early Quenya, first appearing as ᴱQ. yunqe in the Early Qenya Grammar from the 1920s (PE14/82), so it is likely that the √RASAT “twelve” co-existed with ✶yūnekē when that root was introduced in the 1930s. In currently published materials, rasta only appears as a suffix in yurasta “twenty four” (twice-twelve) in a discussion of the Elvish duodecimal (base twelve) counting system, also probably from the 1930s (PE14/17). It may be that ✶yūnekē was used for the general word for “twelve”, while √RASAT was used for “groups of twelve” as part of this duodecimal system. If so, it may have only been used as a suffix in grouping words in this counting system.
ᴹ√RAT “go in a line (as a road), walk”
This root first appeared as ᴹ√RAT “walk” in The Etymologies of the 1930s with derivatives like N. râd “path, track”, N. ostrad “street”, N. rath “course, river-bed”, and N. rant “lode, vein”, the last with the meaning Ilk. rant “flow, course of river” in Ilkorin (Ety/RAT). Hints of this root can be seen as early as the Gnomish Lexicon of the 1910s in words like G. rada “track, path, way” with primitive form rad·, probably actually *ᴱ√RATA (GL/64).
ᴹ√RATA reappeared on a rejected page of roots in the Quenya Verbal System from the 1940s with the gloss “go in a line (as a road)” (PE22/127). Above it Tolkien wrote “usually of animals/or two feet is {SRATA}”, perhaps indicating Tolkien was divorcing this root from the sense “walk”, which in later writings seems to be attributed to the root *√PAT (PE17/34). Tolkien’s continued use of S. rant in names like S. Adurant “Double Stream” and Celebrant “Silverlode” indicate the ongoing validity of √RAT, but in Nomenclature of the Lord of the Rings from the late 1960s, Tolkien clarified that “lode” in “Silverlode” meant “course, water-channel” (RC/775), so it may be that the Sindarin word rant inherited the Ilkorin meaning “flow, course of river” as much as the Noldorin meaning “lode, vein”.
S. rath also appeared regularly in Tolkien’s later writing, but usually with the gloss “street” (LotR/768, 826) = N. ostrad rather than its Noldorin meaning “course, river-bed”; only the name S. Rathlóriel “Golden-bed” (S/235) seems to preserve the 1930s meaning of rath = “river-bed”. Nevertheless, Tolkien may have intended for S. rath “street” to still be derived from √RAT, though in 1969 notes he derived it from √RATH instead (UT/255); see that entry for details. Likewise S. athrad “ford” continued to appear in Tolkien’s later writings, but its later etymology is muddled as well, in one place derived as a gerund (“crossing”) of the verb S. athra- “to cross” (PE17/14). N. râd “path” is only clearly evident in the word S. imrad “a path or pass between mountains, hills or trackless forest” from the late 1960s (VT47/14).
Neo-Eldarin: Despite the mixed etymologies of its derived words in later writings, I think it best to consider ᴹ√RAT as still valid, though perhaps with a meaning “go in a line or on a course” rather than as “walk”.
√RATH, √RET, ᴹ√RAP “climb”
Tolkien gave various roots meaning “climb” in the 1940s, 50s and 60s. The first of these was ᴹ√RAP “climb (as a tree by clinging[?])” on a rejected page of verbal roots in the Quenya Verbal System of the 1940s, with a single derived (verb?) form ᴹQ. rampa (PE22/127). In 1957 Notes on Names (NN) Tolkien gave √RET “climb” as part of an explanation of the name S. Orodreth, translated there as “Mountaineer (= *mountain climber)”, the only place Tolkien ever explained this name (PE17/182). The root √RATH “climb” appeared in a 1969 explanation of the name S. Amroth as “upclimber, high climber”; in this same note he also derived S. rath “street” from this root, as it “applied to all the longer roadways and streets of Minas Tirith, nearly all of which were on an incline” (UT/245, 255 note #16). Finally √RATH appeared in 1970 green-ink revisions to the Outline of Phonology, but there the root was unglossed (PE19/89 and note #101).
Neo-Eldarin: For purposes of Neo-Eldarin, I think √RET is probably the best choice for “climb”, since √RATH would have too many conflicts with the various derivatives of ᴹ√RAT “walk, go in a line”.
ᴹ√RAUTA “metal”
A root in The Etymologies of the 1930s, first glossed “copper” and then “metal”, with derivatives ᴹQ. rauta/N. rhaud “metal” and serving mainly to explain names like N. Angrod and N. Finrod (Ety/RAUTĀ). It might have reappeared in the name S. Rodëol “metal of Eöl” from later Silmarillion drafts (WJ/322), but by 1957 Tolkien was explain the final element of S. Finrod as S. raud “noble” (PE17/49, 118), so I think it likely that ᴹ√RAUTA “metal” was abandoned.
Neo-Eldarin: For purposes of Neo-Eldarin, I’d use Q. tinco (LotR/1122) and its cognate [N.] tinc (Ety/TINKŌ) for “metal” words. For “copper” I’d use derivatives of √(U)RUS (VT41/10).
ᴹ√RAW “*roar”
Tolkien used similar Elvish words for “lion” for much of his life. Lion words were connected to a pair of related roots in the Qenya Lexicon of the 1910s: ᴱ√RAVA or ᴱ√RAẆA (QL/79). The first of these had derivatives like ᴱQ. rauna “prey” and ᴱQ. rauta- “chase, hunt, pursue” as well as ᴱQ. rau “lion”. The second had derivatives like ᴱQ. rauka “swift” and ᴱQ. rawa- “run, chase”. In the contemporaneous Gnomish Lexicon, words like G. raug “rushing, swift” were probably derived from ᴱ√RAẆA and words like G. rau “lion” and G. raust “hunt, chase” were probably derived from ᴱ√RAVA (GL/65).
In The Etymologies of the 1930s Tolkien gave only the root ᴹ√RAW (unglossed) with derivatives like ᴹQ. rá/N. rhaw “lion” (Ety/RAW). Hints of this root are also seen in later words like S. Rauros “Roaring Spray” (RC/327) and Q. rávë “roaring noise” (MC/223), indicating that the root meaning might be “*roar”. As for “lion”, in the Outline of Phonology (OP2) Tolkien have two distinct primitive forms: ✶rāw “lion” and ✶raurō “roarer” (PE19/99, 102, 104).
√RAY¹ “net, knit, contrive network or lace, involve in a network, enlace”; √RAY² “smile”
A root appearing in pair notes from around 1969 glossed “net; knit (contrive a network); catch, involve (in a net)” (PE22/159) and “net, knit, contrive network or lace; {catch,} involve in a network, enlace” (VT42/12), in both notes serving as an explanation for the name S. Gilraen. Both notes have a similar set of derivatives such as Q. raima/S. raef “net” and Q. raina/S. raen “netted, enlaced”. In the second note this root was compared to √WIG and √REB/REM, the former used of weaving in general as opposed to √RAY used of lace-work with a single thread, and the latter a verbal root having to with netting via traps and snares such as by hunters and fisherman. This comparison to √REB/REM was probably related to the deletion of the gloss “catch” from the second note.
These notes probably replaced another etymology of the name S. Gilraen in notes from the late 1950s where √RAY² was glossed “smile” and “show pleasure or favour in facial expression” with derivatives like Q. raina/S. raen “smiling” and contrasted with the root √LALA “laugh” (PE17/182; VT44/35). This in turn might have been connected to Q. rainë “*peace, good will” in Quenya prayers from the 1950s (VT44/34).
Neo-Eldarin: It is very unlikely Tolkien even considered both √RAY “knit” and √RAY “smile” to be part of Elvish at the same time. Unfortunately, the latter is the most popular source of “smile” words in Neo-Eldarin; the only other option is the very early root ᴱ√MIRI or ᴱ√MṚT͡YṚ “smile”, whose derivatives clashes with the later and very well-known root √MIR “precious” used for jewels. It is therefore difficult to eliminate √RAY “smile” in favor of √RAY “knit”, itself with a large number of useful derivatives. For now, I recommend keeping both as part of Neo-Eldarin, even though they were probably incompatible in Tolkien’s mind.